“The fluoride in our water is completely safe,” I heard myself saying with practiced confidence.
As has become increasingly common, I was debating the health of America’s children with an inquisitive family during a routine emergency department consultation. As a physician, I felt an instinctive need to defend medical orthodoxy.
“Is it? How can you be so sure?” The patient’s father asked, his skepticism cutting through my certainty.
The question lingered uncomfortably. No one becomes a pediatrician expecting controversy, but somewhere along the way, our field had become a battleground. Everything is being questioned — from vaccines to nutrition guidelines and yes, even water fluoridation. My initial reaction was to dismiss these doubts, but a nagging feeling remained: how can I promise complete safety?
The family’s skepticism forced me to confront an essential distinction in modern medicine: the difference between no harm and no known harm. Although this may seem like a sleight of hand — a magician’s trick — it represents a fundamental shift in how we calculate risk.
Consider the evidence. A recent meta-analysis in JAMA Pediatrics examining 74 cross-sectional and prospective cohort studies found a significant inverse association between fluoride exposure and children’s IQ scores. However, this was only observed when levels were above 1.5 mg/L — higher than the CDC’s recommendation of 0.7 mg/L but much lower than the EPA’s maximum contaminant level standard of 4.0 mg/L. Although I confirmed our local levels are around 0.7 mg/L, reading the National Toxicology Program’s recent assessment gave me pause. After extensive review, they concluded there was “insufficient data to determine if the low fluoride level of 0.7 mg/L currently recommended for U.S. community water supplies has a negative effect on children’s IQ.” In other words, they said there is no definitive evidence of harm but also no definitive evidence for safety.
“Based on current evidence, fluoride at our local level likely poses minimal IQ risk, but we need more research to be certain,” is the nuanced reply that I should have given to the family.
The reality is more complex than water levels alone suggest. We lack high-quality studies that account for total fluoride exposure. Children are exposed to fluoride not just from water, but from various sources, including infant formula and, most significantly, toothpaste. When parents follow recommendations to brush their child’s teeth twice daily after tooth eruption, infants and toddlers often swallow some of the toothpaste. This cumulative exposure remains understudied.
The situation becomes even more complex when considering the difficult-to-study prenatal population. A small prospective cohort study of mother-child pairs in Los Angeles found prenatal fluoride exposure was associated with increased risk of neurobehavioral problems in the children. While these findings need further validation, they challenge our assumption of safety, and highlight a crucial blind spot in our understanding. They exist in that uncomfortable space between “no harm” and “no known harm.”
“However, there are significant gaps in our understanding,” I should have added for the family.
Before I left the room, the father asked another important question, “Is fluoride in water even needed anymore?” This question cuts to the heart of risk-benefit analysis in modern medicine. While water fluoridation was undoubtedly revolutionary in preventing dental caries historically, its benefits should be reevaluated in an era of widespread fluorinated toothpaste use and regular dental care. Instead of a universal fluoridation recommendation, perhaps the recommendations should vary by locality and take into account local factors like socioeconomic status and access to regular dental visits, as some studies suggest. Moreover, a reassessment of risk to inform local-level recommendations could theoretically work well for an intervention like fluoridation, as opposed to an intervention like vaccination, because fluoridation deals with individual risk as opposed to population risk or herd immunity.
Importantly, there are substantial risks with fluoride removal, especially in vulnerable populations, while the potential harm of fluoridated water is likely small. So, updated recommendations should balance not only the likelihood of harm and benefit, but also the effect size in both directions.
One positive outcome of the post-COVID 19 era is increased public engagement with public health interventions — but this is only a positive if we allowed it to be. It requires us, as medical professionals, to be more nuanced in our communication. Over-promising safety while understating uncertainty is a sure way to lose public confidence.
This brings us to the foundational principle of medicine: first, do no harm. Gradually, we’ve shifted from “prove it’s safe” to “prove it’s harmful,” inverting the burden of proof in our public health interventions. While this reversal might be justified when benefits clearly outweigh potential risks, improvements in modern dental care make this benefit calculation less clear for water fluoridation. The question isn’t simply whether fluoridation offers benefits, but whether its continued universal application meets our ethical obligation to minimize harm, no matter how small.
“You raise important points,” I wish I’d told the father. “The recommendation for universal fluoridation should be re-evaluated.”
Dua Hassan, MD, MPH, is an instructor in pediatrics at Stanford University School of Medicine.
Please enable JavaScript to view the